Somatic mutations, chromosomal defects and epigenetic changes constitute the key pathogenic defects in myelodysplastic syndrome (MDS). Recent scientific advances in molecular technologies have led to the discovery of new classes of recurrent lesions and the identification of novel molecular pathways of oncogenesis or mutations associated with specific pathomorphologic features. This proposal focuses on a newly discovered novel class of mutations affecting spliceosomal genes and among them U2AF1. Highly recurrent, heterozygous missense mutations affecting 2 zinc finger domains in this gene are frequent in MDS and AML where they are prognostic for accelerated progression and poor survival. Because spliceosomal mutations appear to be particularly frequent in certain forms of MDS, this disease will serve as a model for investigation of mechanisms by which spliceosomal defects mediate oncogenic effects. Our proposal is based on the hypothesis that defects in spliceosomal genes due to somatic mutations lead to specific types of mis- splicing of distinct combinations of tumor suppressor genes (TSG) and therefore ultimately result in pathogenetic consequences similar to those produced by direct lesions to these TSG. Hence, spliceosomal mutations may phenocopy consequences of other genomic defects. On the molecular level, U2AF1 mutations result in change of function through differential exclusion of specific splice site sequences and thereby result in creation of specific mis-splicing patterns. In this project, w will investigate the recurrent mis-splicing patterns and the structure-function relationship of the splicing defects due to U2AF1 mutations in MDS and identify exons affected by mis-splicing. We will determine whether splicing dysfunction and the aberrant splicing patterns observed in patients can be recapitulated in engineered model cell lines. We will also compare the RNA binding specificities of purified recombinant wild type and mutant U2AF as well as the effects on in vitro splicing of mispliced target genes. Furthermore, we will restore normal splicing in cells y introducing decoy RNAs with binding sites for the mutant and investigate whether the inhibition of PP1/PP2 phosphatases can improve spliceosomal function. Finally, we will analyze the clinical consequences of U2AF1 mutations. Mutation-associated phenotypes and outcomes will be compared to those seen in patients without U2AF1 mutations but with the haploinsufficient expression/hypomorphic function of downstream genes found to be otherwise affected in by U2AF1 defects.